Speedup likelihood calc

[amn3142]

2 main changes:

1. This speeds up the calculation: first if we don't have imaging data it doesn't do the densitysamples estimate for the array of ones, second, it only calculates the imaging data likelihood one time and not n-bootstraps times.

2. Now can take a covariance matrix. Sending an email for how to estimate the covariance matrix. Haven't tested that this works properly TBH.

[dangilman]

@amn3142 I just push some changes to the main branch that implement the features in the pull request, let me know what you think. The compute_likelihoods function now returns the imaging likelihood, flux ratio likelihood, and then (parameters, weights), meaning it returns one fewer argument than before

[amn3142]

Exciting! Questions: 1) Do you still want the if statement on line 87? 2) If you go back to the bootstrap method, is there any benefit for computing pdf_imgdata multiple times? If not, I would take it out of the for loop. The output makes sense to me. Anna On Jul 23, 2025, at 7:32 PM, Daniel Gilman ***@***.***> wrote: [https://avatars.githubusercontent.com/u/15677070?s=20&v=4]dangilman left a comment (dangilman/samana#16)<#16 (comment)> @amn3142<https://github.com/amn3142> I just push some changes to the main branch that implement the features in the pull request, let me know what you think. The compute_likelihoods function now returns the imaging likelihood, flux ratio likelihood, and then (parameters, weights), meaning it returns one fewer argument than before — Reply to this email directly, view it on GitHub<#16 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AEA6XOUGOCUGWUQJ5FDEGJT3KBAVFAVCNFSM6AAAAACCEG2HG6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTCMJRG42TCMRSG4>. You are receiving this because you were mentioned.Message ID: ***@***.***>

…

-------------------------------------------- Anna M. Nierenberg Assistant Professor Department of Physics, UC Merced annanierenberg.com

[dangilman]

@amn3142 I think it's ok to keep the if statement there for backwards compatibility. And the imaging data stuff is inside the if boostrap_index == 0: statement, so it will only get computed once on the first pass through the for loop

[amn3142]

Got it!! Sorry I missed that. On Jul 23, 2025, at 7:44 PM, Daniel Gilman ***@***.***> wrote: [https://avatars.githubusercontent.com/u/15677070?s=20&v=4]dangilman left a comment (dangilman/samana#16)<#16 (comment)> @amn3142<https://github.com/amn3142> I think it's ok to keep the if statement there for backwards compatibility. And the imaging data stuff is inside the if boostrap_index == 0: statement, so it will only get computed once on the first pass through the for loop — Reply to this email directly, view it on GitHub<#16 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AEA6XOWVSNSTCPXW2UZAO3D3KBB73AVCNFSM6AAAAACCEG2HG6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTCMJRG43DMMZUGY>. You are receiving this because you were mentioned.Message ID: ***@***.***>

…

-------------------------------------------- Anna M. Nierenberg Assistant Professor Department of Physics, UC Merced annanierenberg.com